the architecture of tetrahymena telomerase...

ARTICLEdoi:10.1038/nature12062

The architecture of Tetrahymenatelomerase holoenzymeJiansen Jiang1,2,3*, Edward J. Miracco2*, Kyungah Hong4, Barbara Eckert4, Henry Chan2, Darian D. Cash2, Bosun Min4,Z. Hong Zhou1,3, Kathleen Collins4 & Juli Feigon2,3

Telomerase adds telomeric repeats to chromosome ends using an internal RNA template and a specialized telomerasereverse transcriptase (TERT), thereby maintaining genome integrity. Little is known about the physical relationshipsamong protein and RNA subunits within a biologically functional holoenzyme. Here we describe the architecture ofTetrahymena thermophila telomerase holoenzyme determined by electron microscopy. Six of the seven proteins and theTERT-binding regions of telomerase RNA (TER) have been localized by affinity labelling. Fitting with high-resolutionstructures reveals the organization of TERT, TER and p65 in the ribonucleoprotein (RNP) catalytic core. p50 has anunanticipated role as a hub between the RNP catalytic core, p75–p19–p45 subcomplex, and the DNA-binding Teb1. Acomplete in vitro holoenzyme reconstitution assigns function to these interactions in processive telomeric repeatsynthesis. These studies provide the first view of the extensive network of subunit associations necessary fortelomerase holoenzyme assembly and physiological function.

Telomerase is a unique endogenous eukaryotic reverse transcriptase(RT) required for maintenance of linear chromosome ends and is ahighly regulated determinant of cellular ageing, stem cell renewal andtumorigenesis1,2. TER contains a region of sequence complementarityto the telomeric repeat that is used as a template for addition ofsuccessive repeats to the 39 end of chromosomes, for example additionof the repeat TTAGGG in humans. The specialized telomerase cata-lytic cycle of telomeric repeat synthesis has been extensively investi-gated for the human enzyme and biochemically similar enzymes frommodel organisms such as the telomere-rich ciliate Tetrahymena ther-mophila3. Whereas only TERT and TER are required for telomerasecatalytic activity in vitro, the physiologically functional holoenzyme isa multi-subunit RNP4,5. Affinity purification, mass spectrometry andsubunit tagging assays have identified eight telomerase-specific, reci-procally co-purifying subunits of Tetrahymena telomerase holo-enzyme, each essential for telomere length maintenance6,7 (Fig. 1a).Three of these subunits are required for in vivo assembly of a cataly-tically active RNP: TERT, TER and the La-family protein p657. Thep65 La and RRM1 domains increase p65 affinity for TER, but only thecarboxy-terminal xRRM2 domain (Fig. 1a) is critical for the TERfolding that enhances TERT RNP assembly in vitro and in vivo8–11.Two TER regions, loop 4 (L4) next to the p65 xRRM2 interaction siteand the TERT high-affinity binding element (TBE) 59 to the template(Fig. 1a), bind to the TERT high-affinity RNA binding domain(TRBD)12–14. The TERT TRBD, RT domain and C-terminal extension(CTE) form a ring that encircles an active site cavity, with a fourthTERT amino-terminal (TEN) domain (Fig. 1a) positioned in a TER-dependent but otherwise unknown location15–19.

Little is known about the configuration and roles of the five addi-tional Tetrahymena telomerase holoenzyme proteins required for invivo telomere elongation (p75, p50, p45, p19 and Teb1; Fig. 1a); onlyTeb1 has a known domain structure. Like the paralogous large sub-unit of the single-stranded DNA (ssDNA) binding factor replicationprotein A, Teb1 has four OB-fold domains, NABC (Fig. 1a); A and B

bind ssDNA with high affinity and C is necessary for holoenzymeassociation6,20,21. Teb1 is required for the particularly high DNA prod-uct interaction stability of an endogenously assembled Tetrahymenatelomerase holoenzyme, evident as high repeat addition processivity(RAP) in vitro6,20,21. The holoenzyme subunits p75, p19 and p45, heredesignated 7-1-4, form a subcomplex that remains assembled uponmicrococcal nuclease-induced dissociation of the other holoenzymesubunits in vitro6,21. Little is known about p50, which seems sub-stoichiometric on silver-stained gels of affinity purified holoenzyme6.

Here we report the native electron microscopy structure and acomplete in vitro reconstitution of Tetrahymena telomerase holo-enzyme. This first physical and functional network architecture of atelomerase holoenzyme provides unprecedented detail about thestructure of the RNP catalytic core and reveals the organization ofholoenzyme subunits that confer processivity and bridge telomeraseto telomeres.

Overall structure and localization of subunitsTetrahymena telomerase holoenzymes were prepared for electronmicroscopy (Supplementary Fig. 1) by affinity purification6 from 10different strains bearing N- or C-terminal 33Flag (F) and tandemprotein A (ZZ) tags (ZZF or FZZ, respectively) on TERT or otherholoenzyme subunits (Supplementary Table). Comparison of theclass averages from negative staining electron microscopy images oftelomerase purified using TERT–FZZ (TERT-F telomerase, whichresults from cleavage of the ZZ tag) to those from cryo-electronmicroscopy images (Fig. 1b and Supplementary Fig. 2a, b) shows thatthe negative stain did not change the structure within experimentalresolution. In addition to the predominant (‘stable’) conformation ofthe complete holoenzyme particle, constituting 31% of the total part-icles, the class averages show additional particle subpopulations ofvarying conformation and subunit composition (described belowand Supplementary Fig. 2). The class averages from both negativestaining electron microscopy images and cryo-electron microscopy

*These authors contributed equally to this work.

1Department of Microbiology, Immunology and Molecular Genetics, University of California, Los Angeles, California 90095, USA. 2Department of Chemistry and Biochemistry, University of California, LosAngeles, California 90095, USA. 3California Nanosystems Institute, University of California, Los Angeles, California 90095, USA. 4Department of Molecular and Cell Biology, University of California, Berkeley,California 94720, USA.

0 0 M O N T H 2 0 1 3 | V O L 0 0 0 | N A T U R E | 1

Macmillan Publishers Limited. All rights reserved©2013

www.nature.com/doifinder/10.1038/nature12062

images show a strongly preferred orientation (Supplementary Fig. 2).To overcome the problems of preferred orientation and structuralvariability, the electron microscopy image acquisition and three-dimensional (3D) reconstructions were carried out using an automatedrandom conical tilt (RCT) method22 (Supplementary Fig. 3). The 3Dstructures of TERT–F and Teb1–F telomerase holoenzymes are indis-tinguishable (Supplementary Fig. 4a), and Teb1–FZZ gave higher yieldsof holoenzyme. A 3D structure (Fig. 1c) of the ,500 kDa Teb1–Ftelomerase holoenzyme from 3D reconstruction of 2,220 particles witha resolution of ,25 A (Supplementary Fig. 4b) demonstrates well-ordered holoenzyme density occupying about 200 3 150 3 80 A.

To locate each protein subunit within the overall holoenzyme struc-ture, telomerase was purified from strains harbouring FZZ-taggedTERT, p75, p65, p45, p19 or Teb1 in place of the correspondinguntagged protein6. During purification, the ZZ portion of the tag wasremoved by proteolytic cleavage, leaving only the short F tag that waslabelled using the antigen-binding fragment (Fab) derived from mono-clonal anti-Flag antibody (see Methods). Each tagged protein can bindup to three Fabs, as illustrated in some of the class averages and 3D maps(Fig. 1c and Supplementary Fig. 5), which pinpoint to a single spot andlocalize the binding site of Fab unambiguously. Affinity purification ofthe biologically functional C-terminally tagged p50 (p50–FZZ) does notenrich sufficient holoenzyme for silver staining detection in SDS–PAGE, probably due to proteolysis between the C-terminal tag locationand the region of p50 required for holoenzyme assembly6. Instead weused N-terminally ZZF-tagged p50 (ZZF–p50) expressed in partialreplacement of the endogenous locus, which did purify holoenzymeto homogeneity (Supplementary Fig. 6). All holoenzyme protein sub-units were localized by Fab labelling except p45, which was identified byprocess of elimination (see Methods). In addition, we affinity purifiedtelomerase from a strain with biologically functional tagged TER har-bouring a small hairpin tag (MS2hp), which is recognized by the MS2coat protein (MS2cp), appended to TER stem 2 (S2)23 (Fig. 1c andSupplementary Fig. 5g). The MS2cp dimer bound to this tag appearsas extra density in the class averages and thus provides the approximatelocation of the TBE 59 to the template (Fig. 1a).

The positions of the TERT and Teb1 C termini and the p50 Nterminus could be precisely mapped in the 3D reconstructions of

the respective Fab-labelled telomerase particles (Supplementary Fig.5a–c), whereas a less precise location of the C termini of p65, p75 andp19 (which due to low purification yield did not give sufficient part-icles for RCT data collection) was obtained from the class averages(Supplementary Table, Fig. 1c and Supplementary Fig. 5d–f).Approximate subunit boundaries (schematized in Fig. 1d) could bemodelled after fitting the holoenzyme density with known TERT15–18,TER8,24–26, p658 and Teb120 high-resolution domain structures com-bined with comparisons of class averages and 3D reconstructionsacross different complexes as described below. These data reveal that,in the ‘front’ view in Fig. 1c (schematized in Fig. 1d) with p65 at thebottom, TERT occupies the centre of the particle and 7-1-4 subunitsform the top. Teb1 projects from the middle layer to the right. The p50subunit is also part of the middle layer, networked between TERT,7-1-4 and Teb1 (Fig. 1c, d). Numerous inter-subunit appositionsgenerate an intricate and highly contoured surface.

The structure of the RNP catalytic coreBased on the determined subunit locations we fit the crystal and NMRstructures of TER and protein domains in the electron microscopydensity map of Teb1–F telomerase to generate a model of the entireRNP catalytic core (Fig. 2a and Supplementary Video). The RT andCTE domains of Tetrahymena TERT were homology modelled27 fromthe Tribolium castaneum TERT crystal structure18 and combined withthe crystal structures of Tetrahymena partial TRBD15 and TEN16

domains as described in Methods. The modelled TERT TRBD-RT-CTE fits in only one orientation (Fig. 2a, b and Supplementary Fig. 7a,b). The position of the TERT CTE is consistent with the location of theC terminus identified in the 3D reconstruction of TERT–F–Fab tel-omerase (Fig. 1c and Supplementary Fig. 5a)17. The TEN domain wasplaced into density remaining after determination of subunit bound-aries of the adjacent Teb1C and p50, and oriented based on thehomology model of human TERT28. The modelled TetrahymenaTERT TRBD interacts with the CTE, which is consistent with theTribolium TERT crystal structures17,18 and isolated human TERTdomain interactions19. There is electron microscopy density linkingthe TEN and CTE regions and the TEN and TRBD regions, whichcould correspond to the ,70 amino acids of potentially unstructured

Negative staining CryoEM

90°

Front view Sideview

p65

N

TRBD

TER

p45 p19

TERT

A B C

TEN

N La RRM1 xRRM2

RT CTE

p75

7-1-4

Teb1

p50

MS2hp–MS2cp

TERT–F–Fab

p65–F–Fab

Fab–F–p50

Teb1–F–Fabp19–F–Fabp75–F–Fab

p65/TER

Teb1CTERT/TER

p50

p45p19

p75

Teb1B

S1

SL2

t/PK

TBE

TRE

PK

SL4

5′

3′UUUU

Te

mplate

a b

c

d

Figure 1 | Electron microscopy reconstruction ofTetrahymena telomerase holoenzyme andsubunit localization. a, Holoenzyme subunits anddomains (top) and TER secondary structure(below). b, Representative class averages ofnegative staining electron microscopy and cryo-electron microscopy images of TERT–Ftelomerase. c, 3D reconstruction of Teb1–Ftelomerase (front and side views) and classaverages of affinity-labelled telomerase particles.Lines with circle heads indicate attachment point ofFab (red arrows) and MS2cp (white arrow). Side-lengths of class averages in this and subsequentfigures are 350 A. d, Subunit schematic (frontview).

RESEARCH ARTICLE

2 | N A T U R E | V O L 0 0 0 | 0 0 M O N T H 2 0 1 3


TEN-TRBD linker29,30 and adjacent TRBD sequence missing fromhigh-resolution structures15,16, which crosslinks to ssDNA and improvestemplate boundary definition19,31.

The C terminus of p65 is at the bottom of the particle, below TERT(Fig. 1d and Supplementary Fig. 5d). In the class averages and 3Dreconstructions we observed particle subpopulations lacking part ofthe density at the bottom left of the particle (Fig. 2c and SupplementaryFigs 2, 3c and 5d), which must arise from loss of p65 N-terminaldomain(s) due to partial proteolysis of this subunit common duringholoenzyme purification6 and/or consequent loss of positional con-straints on the remaining La and RRM1 domains. The structure of thep65 C-terminal domain xRRM2–stem 4 (S4) complex8 could thereforebe localized and fit to the density present in all the 3D reconstructions(Fig. 2a, d and Supplementary Fig. 7c, d). The remaining parts ofS1–SL4 and p65 La and RRM1 could then be modelled (see Methods).The p65 N-terminal domain (near S1) and the long b2–b3 loop withinxRRM2 that was deleted in the crystal structure could occupy theremaining unmodelled density (Fig. 2a)8,26. Together these fittingsprovide the overall topology of p65–TER interaction (Fig. 2a, d).

Tetrahymena TER contains two major domains, the template/pseudoknot (t/PK) and SL4, which are connected by S1 (Fig. 1a).Starting with the defined locations of the template in the active siteof TERT18, S2 by MS2cp labelling, and SL4 in complex with p65xRRM28 (Fig. 2d, magenta), and considering topological restrictionsbased on the length of the single-stranded regions of TER, a modelstructure of the PK, and S1, we traced a potential trajectory of theentire TER (Fig. 2a, d, e) fit into the remaining electron microscopydensity. The template recognition element (TRE) seems to be close tothe TEN domain, as implicated biochemically32, and the bottom of thePK is close to L4 (Fig. 2a, d). The locations of TER elements in themodel are consistent with the large body of biochemical data on TERstructure and function4,5,33,34. Of particular significance are the well-determined locations of the two TRBD binding elements, the TBE andL49,10,13,32, which we find bind to two distinct regions of the TRBD, nearthe top left and bottom of the TERT ring, respectively (Fig. 2b, d, e).

Remarkably, only the apical loop of SL4 connects to the densityassigned to TERT, resulting in a U-shape at the bottom of the holoen-zyme formed by TER S1/SL4 bound by p65. (Fig. 2a, d, e). The inter-action of distal SL4 with TERT explains how it can stimulatetelomerase activity when added to the t/PK in trans35,36 and howbinding of p65 xRRM2 to S4 mediates hierarchical assembly ofp65–TER with TERT8–10. The electron microscopy structure and

model reveal that the TRBD bridges the TBE and L4, which are,40 A apart. In the model, distal SL4 approaches the 39 end of thePK, consistent with reported high fluorescence resonance energytransfer (FRET) between L4 and an internal PK position37. Althoughdetailed interactions await a high-resolution structure, in the model theTBE contacts the TRBD in the T-pocket and CP-motif region (T-CPpocket) (Fig. 2f), which is consistent with biochemical data12,14,15, anddistal SL4 is flanked by the CTE and TRBD (Fig. 2g), potentially con-tributing to folding of TERT around TER.

p50 anchors the accessory proteinsThe catalytic activity associated with telomerase purified using ZZF–p50 has low RAP, in contrast with the high-RAP activity enriched bythe much lower yield of holoenzyme purified by p50–FZZ (Fig. 3a).The Fab labelling showed that the N terminus of p50 and the Cterminus of Teb1 are relatively close together (Fig. 1c and Sup-plementary Fig. 5b, c). Comparison of the class averages of F–p50and Teb1–F telomerase shows that all F–p50 telomerase particles lackdensity for Teb1 and reveals the boundary of Teb1 by difference map(Fig. 3b). The boundary of p50 was further defined by comparing classaverages from a small subclass of particles containing only the RNPcatalytic core, which was observed only in the MS2cp-bound MS2hptelomerase, with class averages containing p50 (Fig. 3c). Super-positions of 3D reconstructions of Fab–F–p50 and Teb1–F telomeraseon TERT–F telomerase lacking Teb1 (see below) show that Fab boundto the N terminus of F–p50 occupies the same space as Teb1 (Fig. 3d).Together, these results identify the locations of Teb1 and p50, showthat p50 interacts directly with the RNP catalytic core, and reveal thatTeb1 is in close proximity to the N terminus of p50 and to TERT. Theholoenzyme structure therefore explains the low-RAP activity of puri-fied N-terminally tagged p50 (Fig. 3a), because enzyme purification bythe N-terminal ZZF–p50 tag disrupts Teb1 binding. Beyond densityassigned to the RNP catalytic core, Teb1, and p50, the remainingdensity in the top part of the 3D map is occupied by the 7-1-4 sub-complex. Taken together, 3D reconstructions reveal a central locationfor p50 in holoenzyme architecture, as an interaction hub betweenTERT, 7-1-4 and Teb1 (schematized in Fig. 3f).

Assigning function to subunit architectureIn parallel with structural analysis, we developed a method for holoen-zyme reconstitution from entirely recombinant subunits. RNP catalytic

d

p65–F

p65 N-term (–)

e

fTeb1C

p65

TERT

TER

TEN

SL2Template

SL4

PK

TEN

TRBD

xRRM2

RRM1

La

SL2

TBE

S4TRBD

RT

TRBD

T-CPpocket

TemplateTRE

TBE

L4

a b c

60°

S1

CTE

CTE

xRRM25′3′

SL2 Template

SL4

PK

TRE

S1

TRBD

p65N

p65 aa 413–459

g

CTECTE

RTRT

TRBDTRBD

CTE

RT

TRBD

Figure 2 | Structure of the RNP catalytic core.a, 3D reconstruction of Teb1–F telomerase withTERT, p65 and TER (black), plus Teb1C modelledinto the electron microscopy density. The dashedline indicates the top boundary of TERT/TER.b, Zoomed and rotated view of a showing TERTdomains TEN, CTE, TRBD and RT, with TERtemplate and essential Mg21 at the active site inmagenta. c, Class averages of p65–F telomerase(top) and 3D reconstructions (bottom) of TERT–Ftelomerase with p65 (left) and p65 missingN-terminal density (right, black arrows). d, TERmodel structure (well determined, magenta;remaining, black) and interactions with TERTTRBD and TEN and p65 La, RRM1, and xRRM2domains. e, Secondary structure schematic of TERwith TRBD. f, Modelled interaction between TBE(pink) and TRBD T-CP pocket. g, Modelledinteractions of distal SL4 with CTE and bottom ofTRBD. In f, g, TRBD is shown as GRASP surface.

ARTICLE RESEARCH

0 0 M O N T H 2 0 1 3 | V O L 0 0 0 | N A T U R E | 3


core with TERT–F was preassembled in rabbit reticulocyte lysate (RRL)and then combined with RRL-expressed p50 and/or 7-1-4 and/or bac-terially expressed Teb1BC (see Methods)21 (Fig. 3g). Direct primer-extension assays showed that addition of 7-1-4 or Teb1 alone did notalter the low-RAP product synthesis profile of the RNP catalytic core,but addition of p50 alone stimulated enzyme activity and the synthesisof multi-repeat products (Fig. 3g, lane 3). For p50-containing enzymes,7-1-4 alone increased the amount of product (Fig. 3g, lane 5), Teb1alone increased product length (lane 7), and their combination intocomplete holoenzyme was synergistic for increased activity level andlong product synthesis (lane 8). These in vitro biochemical activities ofp50, 7-1-4 and Teb1 were evident within a 5-min reaction time thatallowed all product DNAs to be resolved by gel electrophoresis (Fig. 3g)or with the additional time and concentration of dGTP that supportedvery long product synthesis (Supplementary Fig. 8). The p50-dependentactivity of 7-1-4 and Teb1 explains the presence of p50 in virtually all ofthe electron microscopy structures.

Previous reconstitution assays combined RRL-assembled recom-binant RNP catalytic core, bacterially expressed Teb1 and endogen-ously assembled Tetrahymena proteins that remained bound to p45–Fafter micrococcal nuclease treatment of purified holoenzyme20,21,38. Wefound that the 7-1-4 complex isolated by this method does not com-pletely dissociate p50, which as a full-length protein co-migrates withp45–F (Supplementary Fig. 6). The presence of residual p50 accountsfor the reconstitution of holoenzyme-like catalytic activity by comple-mentation of the RNP catalytic core and Teb1 in previous assays20,21,38.Reconstitutions using entirely recombinant holoenzyme subunitsdemonstrate a p50-dependent influence of 7-1-4 or Teb1 on RNPcatalytic activity (Fig. 3g), consistent with their locations in the electronmicroscopy structures (Fig. 3e, f).

Teb1 domains are positionally flexibleAll telomerase preparations except for those with tagged Teb1 have,5% of particles missing density for Teb1 in the class averages and 3Dreconstructions (Fig. 3e and Supplementary Fig. 2). This is not sur-prising, because Teb1 is sensitive to proteolysis and dissociationduring holoenzyme affinity purification6. To identify specific Teb1domains in the structures, we constructed strains in which TERTwas C-terminally tagged at its endogenous locus with only tandemprotein A domains (TERT–ZZ) and an N-terminally F-tagged Teb1C

or Teb1BC was expressed from an integrated transgene. Holoenzymeparticles isolated by tandem affinity purification of TERT–ZZ/F–Teb1C or TERT–ZZ/F–Teb1BC had the expected high RAP (Fig. 4a)and virtually identical class averages and 3D reconstructions to eachother and to those from purified Teb1–F (Fig. 4b–d and Sup-plementary Fig. 4a). Because Teb1C mediates Teb1 association withother holoenzyme proteins21 and the N-terminal tag on Teb1BC abso-lutely requires the associated holoenzyme to contain Teb1B, we con-clude that the Teb1B domain is positionally flexible and therefore didnot appear in the class averages and 3D reconstructions. Similarly, forTeb1–FZZ purified particles, which could contain all of theTeb1NABC domains, only Teb1C is visible in the majority of classaverages. Teb1C was modelled into the electron microscopy densitymap (Fig. 2a) based on the position of the C terminus identified by Fablabelling and the best fit with the shape of the electron microscopydensity. The orientation around the y axis is not definitive. Weobserved a small set of class averages (containing ,5% of the particles)of F–Teb1BC and Teb1–F telomerases that show a weak extra densityabove Teb1C (Fig. 4c, d), which we assign to Teb1B. The conforma-tional flexibility of Teb1 NAB domains relative to Teb1C is consistentwith structural studies of the paralogous large subunit of replicationprotein A, which has three DNA-binding OB-fold domains thatbecome ordered relative to each other only upon binding to ssDNA39.

7-1-4 subcomplex has multiple orientationsNeither high-resolution structures nor exact biological functions ofthe 7-1-4 subunits are known. Both the electron microscopy struc-tures and previous biochemical data6,21 indicate they constitute a sta-bly assembled subcomplex. Because no class averages lacking anindividual 7-1-4 subunit were observed, inter-subunit boundariescould only be inferred from the class averages and Fab labelling ofthe C termini of p19 and p75, which showed that these two subunitsare close together. In all class averages and 3D reconstructions, themajor class of particles had a 7-1-4 conformation with p75 positionedacross the top of the RNP catalytic core (Fig. 1c and SupplementaryFig. 2) in the stable conformation (Fig. 5). All telomerase samplesshow other positions of 7-1-4 in which it hinges away from theRNP catalytic core while maintaining a physical connection to p50(Fig. 5 and Supplementary Fig. 3c). In these other conformations, theregion of the class averages where 7-1-4 is located appears ‘fuzzier’

g

p50

Teb1

p50

Teb1

p75

p45p19

p50

p75p45p19

a b

c

e

Moc

k

p50–F

F–p50

Teb1–FF–p50

d

f

p50

Teb1

p45p75

p19

TERT

Teb1 (+)

p50 (+)

MS2hp

(1) (2) (3) (4)

7-1-4–

–

+

+

+

Teb1– –

–

––

–

–

+

+

++

+ +

+++

–

––

RC

p50

1 2 3 4 5 6 7 8 Lane

(1) (2) (3) (4)

Figure 3 | p50 anchors TERT, 7-1-4 and Teb1. a, Primer extension assay ofFlag antibody purifications from cell extracts lacking a tagged subunit (mock)or with F–p50 or p50–F. b, Class averages of F–p50 telomerase (1), Teb1–Ftelomerase (2), difference map by subtracting (1) from (2) (seen in 3), and mapof statistically significant (.4 s) regions in the difference map (4). Black arrowpoints to Teb1 density. c, MS2hp telomerase class averages containing MS2cpwithout p50 (1), with p50 (2), difference map (3), and statistically significantregions (4) as in b. White and black arrows point to MS2cp and p50 densities,respectively. 7-1-4 is not seen in these class averages. d, Back view of 3D

reconstructions of TERT–F telomerase lacking Teb1 (gold) overlaid with (top)Fab–F–p50 (grey mesh) and (bottom) Teb1–F telomerase (grey mesh) showingthat Fab (red arrow) occupies the same site as Teb1 (black arrow). e, 3Dreconstructions of TERT–F telomerase lacking 7-1-4 (left), lacking Teb1(middle), and holoenzyme (right). f, Schematic of subunit interactions.g, Primer extension assay of the RNP catalytic core (TERT-TER-p65)reconstituted with additional combinations of 7-1-4, p50, and/or Teb1BC.Reactions were for 5 min. RC is a radiolabelled oligonucleotide added totelomerase products as a precipitation recovery control.

RESEARCH ARTICLE

4 | N A T U R E | V O L 0 0 0 | 0 0 M O N T H 2 0 1 3


than the rest of the particle, consistent with conformational variabilityin this part of telomerase holoenzyme. In ,10% of class averages fromevery purification, including those with tagged p75 or p45, no 7-1-4density is visible, indicating that some 7-1-4 positions are not welldefined relative to the RNP catalytic core (Fig. 3e and Supplemen-tary Fig. 2). Binding of MS2cp to MS2hp holoenzyme seems to favourp75 displacement from the stable position (Fig. 1c and SupplementaryFig. 5g). Analysis of the various conformations indicates that 7-1-4 isrotating as an intact substructure (Fig. 5). Telomerase product DNAcan be ultraviolet crosslinked to p45 as well as the TERT TENdomain31,40, which suggests that at least some orientations of 7-1-4must bring p45 close to DNA.

Implications for holoenzyme assembly and functionTERT and TER comprise a minimal set of telomerase subunits forrepeat synthesis in vitro5,41, but despite extensive efforts the 3Darrangement and molecular interactions between TERT and TERhave not been previously defined. Although divergent in size andsequence, TERs contain a t/PK domain and almost without exceptiona separate activating domain (SL4 in Tetrahymena), each with a TERTbinding site33,42. Here we have unambiguously located two elements ofTetrahymena TER that constitute known binding sites for TERT andmodelled a path of the entire TER. The functional equivalent ofTetrahymena TER distal SL4 in vertebrate TER is P6/P6.1, whichultraviolet crosslinks to the same region of the TRBD where L4 islocated in our model43. Our model structure thus provides insightinto human telomerase TERT–TER interactions and explains theseparable interactions of the TERT binding elements on the t/PKand activating domains of TER.

Using classification and/or 3D reconstruction of 16 differentlylabelled telomerase samples (Supplementary Table), we were able toidentify the subunit arrangement of the holoenzyme. The electronmicroscopy results together with reconstituted enzyme activity assayssuggest a subdivision of holoenzyme functional units: the TERT–TER–p65 catalytic core, p50–Teb1, and 7-1-4. Both structure determinationand activity assays of reconstituted holoenzyme subcomplexes estab-lish a central role for p50, a subunit lacking any predicted domain folds.The electron microscopy data reveal that 7-1-4 is a structural unit withdynamic orientation relative to the rest of the holoenzyme. We suggestthat the positional dynamics of 7-1-4 coordinate telomerase holo-enzyme with additional telomere synthesis or processing activities.

In the model for holoenzyme subunit interactions, the TERT TENdomain, the lower region of p50 and Teb1C have a triangulatedarrangement that suggests coordination by pairwise direct contacts.The TERT TEN domain is required for RAP and has been implicatedin both ssDNA and TER binding5,41,42. Although Teb1 seems to con-tact TERT, subcomplex structures never have Teb1 in the absence ofp50, and reconstituted enzyme activity assays do not show Teb1 func-tion in the absence of p50. The assembly of p50 with the RNP catalyticcore markedly increases processive repeat synthesis, indicating thatp50 stabilizes enzyme association with ssDNA in potential functionalanalogy to yeast Est3 and vertebrate TPP144–48, favouring a productiveTERT TEN domain conformation and/or providing additional ssDNAcontact. Subsequent p50-dependent recruitment of Teb1 furtherenhances activity, which we speculate could occur through interactionof p50 and the Teb1C OB fold. The physical location of p50 determinedby electron microscopy and its functional significance determined byreconstitution assays suggest that p50 could be a critical determinant ofholoenzyme assembly in vivo. This first view of the architecture of acomplete telomerase holoenzyme provides unprecedented opportun-ity to understand the biological mechanisms for coupling of telomeraseto its telomere substrates.

METHODS SUMMARYTetrahymena strain constructions and steps of tag-based affinity purificationwere done as described6 and as in Methods. To label the tagged subunit forelectron microscopy, telomerase particles were first purified using anti-Flag M2antibody resin then bound to rabbit-IgG resin. The telomerase-bound IgG resinwas then incubated with Fab derived from anti-Flag M2 IgG, and elution waseffected by protease cleavage. Negatively stained electron microscopy specimenswere prepared with fresh telomerase samples, stained with 0.8% uranyl formate,and examined with an FEI Tecnai F20 electron microscope operated at 200 kV.Frozen hydrated specimens were prepared using Quantifoil grids and imagedwith an FEI Titan Krios electron microscope operated at 120 kV. The imageprocessing tasks, including image classification and RCT reconstruction, wereperformed as described in Methods.

Telomerase activity assays were performed at room temperature using purifiedtelomerase complexes on Flag antibody resin with standard Tetrahymena holoen-zyme reaction conditions using 0.3mM 32P-labelled dGTP. Holoenzyme recon-stitution used synthetic genes encoding TERT–F, p75, p65, p50, p45 and p19 for

a

55

43

34

26

kDaTERT–ZZ130

Teb1–F

F–Teb1C

F–Teb1BC

b

c

d

– CBC

F–Teb1BC

F–Teb1C

Figure 4 | Contribution of Teb1 domains to holoenzyme structure andactivity. a, Cell extract western blots and two-step purified enzyme primerextension assays of F–Teb1BC (BC, lane 2) and F–Teb1C (C, lane 3) telomerase.Cell extract with TERT–ZZ alone (–, lane 1) is a negative control for specificityof F–Teb1BC and F–Teb1C binding to Flag antibody. b–d, Comparison of classaverages (left column) of F–Teb1C (b), F–Teb1BC (c) and Teb1–F (d)telomerases. Density assigned to Teb1B (white arrows) was seen in ,5% ofparticles, whereas density for Teb1C (black arrow in b) occupies a fixed positionin all particles. Class averages without and with Teb1B density are representedby the upper and lower rows of c and d, respectively. Difference maps (middlecolumn) by subtracting b from the respective class averages and maps ofstatistically significant (.4 s) regions in the difference maps (right column) inc and d show Teb1B density (white arrows).

Stable conformation Other conformations

a

b

p50p50 p50p50 p50p50 p50p50 p50p50

7-1-47-1-47-1-47-1-47-1-47-1-4 7-1-47-1-4 7-1-47-1-4

p50 p50 p50 p50 p50

7-1-47-1-47-1-4 7-1-4 7-1-4

Figure 5 | Positional dynamics of 7-1-4. a, Teb1–F telomerase class averageswith p75 indicated (black arrows). b, 3D reconstructions of Teb1–F telomeraseshowing different positions of 7-1-4.

ARTICLE RESEARCH

0 0 M O N T H 2 0 1 3 | V O L 0 0 0 | N A T U R E | 5


expression in RRL; TER purified following in vitro transcription by T7 RNApolymerase; and N-terminally His6-tagged Teb1BC purified following bacterialexpression21.

Full Methods and any associated references are available in the online version ofthe paper.

Received 22 October 2012; accepted 8 March 2013.

Published online 3 April 2013.

1. Artandi, S. E. & DePinho, R. A. Telomeres and telomerase in cancer. Carcinogenesis31, 9–18 (2010).

2. Armanios, M. & Blackburn, E. H. The telomere syndromes. Nature Rev. Genet. 13,693–704 (2012).

3. Blackburn, E. H., Greider, C. W. & Szostak, J. W. Telomeres and telomerase: Thepath from maize, Tetrahymena and yeast to human cancer and aging. Nature Med.12, 1133–1138 (2006).

4. Egan, E. D. & Collins, K. Biogenesis of telomerase ribonucleoproteins. RNA 18,1747–1759 (2012).

5. Podlevsky, J. D. & Chen, J. J. L. It all comes together at the ends: Telomerasestructure, function, and biogenesis. Mutat. Res. 730, 3–11 (2012).

6. Min, B. & Collins, K. An RPA-related sequence-specific DNA-binding subunit oftelomerase holoenzyme is required for elongation processivity and telomeremaintenance. Mol. Cell 36, 609–619 (2009).

7. Witkin, K. L. & Collins, K. Holoenzyme proteins required for the physiologicalassembly and activity of telomerase. Genes Dev. 18, 1107–1118 (2004).

8. Singh, M. et al.Structural basis for telomerase RNA recognition and RNP assemblyby the holoenzyme La family protein p65. Mol. Cell 47, 16–26 (2012).

9. O’Connor, C. M. & Collins, K. A novel RNA binding domain in Tetrahymenatelomerase p65 initiates hierarchical assembly of telomerase holoenzyme. Mol.Cell. Biol. 26, 2029–2036 (2006).

10. Stone, M. D. et al. Stepwise protein-mediated RNA folding directs assembly oftelomerase ribonucleoprotein. Nature 446, 458–461 (2007).

11. Akiyama, B. M., Loper, J., Najarro, K. & Stone, M. D. The C-terminal domain ofTetrahymena thermophila telomerase holoenzyme protein p65 induces multiplestructural changes in telomerase RNA. RNA 18, 653–660 (2012).

12. Bryan, T. M., Goodrich, K. J. & Cech, T. R. Telomerase RNA bound by protein motifsspecific to telomerase reverse transcriptase. Mol. Cell 6, 493–499 (2000).

13. Lai, C. K., Mitchell, J. R. & Collins, K. RNA binding domain of telomerase reversetranscriptase. Mol. Cell. Biol. 21, 990–1000 (2001).

14. Lai, C. K., Miller, M. C. & Collins, K. Template boundary definition in Tetrahymenatelomerase. Genes Dev. 16, 415–420 (2002).

15. Rouda, S. & Skordalakes, E. Structure of the RNA-binding domain of telomerase:Implications for RNA recognition and binding. Structure 15, 1403–1412 (2007).

16. Jacobs, S. A., Podell, E. R. & Cech, T. R. Crystal structure of the essential N-terminaldomain of telomerase reverse transcriptase. Nature Struct. Mol. Biol. 13, 218–225(2006).

17. Gillis, A. J., Schuller, A. P. & Skordalakes, E. Structure of the Tribolium castaneumtelomerase catalytic subunit TERT. Nature 455, 633–637 (2008).

18. Mitchell, M., Gillis, A., Futahashi, M., Fujiwara, H. & Skordalakes, E. Structural basisfor telomerase catalytic subunit TERT binding to RNA template and telomericDNA. Nature Struct. Mol. Biol. 17, 513–518 (2010).

19. Robart, A. R. & Collins, K. Human telomerase domain interactions capture DNA forTEN domain-dependent processive elongation. Mol. Cell 42, 308–318 (2011).

20. Zeng, Z. et al. Structural basis for Tetrahymena telomerase processivity factor Teb1binding to single-stranded telomeric-repeat DNA. Proc. Natl Acad. Sci. USA 108,20357–20361 (2011).

21. Min, B. & Collins, K. Multiple mechanisms for elongation processivity within thereconstituted Tetrahymena telomerase holoenzyme. J. Biol. Chem. 285,16434–16443 (2010).

22. Radermacher, M., Wagenknecht, T., Verschoor, A. & Frank, J. Three-dimensionalreconstruction from a single-exposure, random conical tilt series applied to the50S ribosomal subunit of Escherichia coli. J. Microsc. 146, 113–136 (1987).

23. Cunningham,D. D. & Collins, K. Biological and biochemical functions of RNA in theTetrahymena telomerase holoenzyme. Mol. Cell. Biol. 25, 4442–4454 (2005).

24. Chen, Y. et al. Structure of stem-loop IV of Tetrahymena telomerase RNA. EMBO J.25, 3156–3166 (2006).

25. Richards, R. J., Theimer, C. A., Finger, L. D. & Feigon, J. Structure of the Tetrahymenathermophila telomerase RNA helix II template boundary element. Nucleic AcidsRes. 34, 816–825 (2006).

26. Richards, R. J. et al. Structural study of elements of Tetrahymena telomerase RNAstem-loop IV domain important for function. RNA 12, 1475–1485 (2006).

27. Arnold, K., Bordoli, L., Kopp, J. & Schwede, T. The SWISS-MODEL workspace: Aweb-based environment for protein structure homology modelling. Bioinformatics22, 195–201 (2006).

28. Steczkiewicz, K. et al. Human telomerase model shows the role of the TEN domainin advancing the double helix for the next polymerization step. Proc. Natl Acad. Sci.USA 108, 9443–9448 (2011).

29. Friedman, K. L. & Cech, T. R. Essential functions of amino-terminal domains in theyeast telomerase catalytic subunit revealed by selection for viable mutants. GenesDev. 13, 2863–2874 (1999).

30. Armbruster, B. N., Banik, S. S. R., Guo, C., Smith, A. C. & Counter, C. M. N-terminaldomainsof thehumantelomerasecatalytic subunit required for enzymeactivity invivo. Mol. Cell. Biol. 21, 7775–7786 (2001).

31. Romi, E. et al. High-resolution physical and functional mapping of the templateadjacent DNA binding site in catalytically active telomerase. Proc. Natl Acad. Sci.USA 104, 8791–8796 (2007).

32. O’Connor, C. M., Lai, C. K. & Collins, K. Two purified domains of telomerase reversetranscriptase reconstitute sequence-specific interactions with RNA. J. Biol. Chem.280, 17533–17539 (2005).

33. Theimer, C. A. & Feigon, J. Structure and function of telomerase RNA. Curr. Opin.Struct. Biol. 16, 307–318 (2006).

34. Cole, D. I. et al. New models of Tetrahymena telomerase RNA from experimentallyderived constraints and modeling. J. Am. Chem. Soc. 134, 20070–20080 (2012).

35. Lai, C. K., Miller, M. C. & Collins, K. Roles for RNA in telomerase nucleotide andrepeat addition processivity. Mol. Cell 11, 1673–1683 (2003).

36. Mason, D. X., Goneska, E. & Greider, C. W. Stem-loop IV of Tetrahymena telomeraseRNA stimulates processivity in trans. Mol. Cell. Biol. 23, 5606–5613 (2003).

37. Wu, J. Y., Stone, M. D. & Zhuang, X. A single-molecule assay for telomerasestructure-function analysis. Nucleic Acids Res. 38, e16 (2010).

38. Eckert, B. & Collins, K. Roles of telomerase reverse transcriptase N-terminaldomain in assembly and activity of Tetrahymena telomerase holoenzyme. J. Biol.Chem. 287, 12805–12814 (2012).

39. Fan, J. & Pavletich, N. P. Structure and conformational change of a replicationprotein A heterotrimer bound to ssDNA. Genes Dev. 26, 2337–2347(2012).

40. Rosenfeld, K. K., Ziv, T., Goldin, S., Glaser, F. & Manor, H. Mapping of DNA bindingsites in the Tetrahymena telomerase holoenzyme proteins by UV cross-linking andmass spectrometry. J. Mol. Biol. 410, 77–92 (2011).

41. Collins, K. Single-stranded DNA repeat synthesis by telomerase. Curr. Opin. Chem.Biol. 15, 643–648 (2011).

42. Blackburn, E. H. & Collins, K. Telomerase: An RNP enzyme synthesizes DNA. ColdSpring Harb. Perspect. Biol. 3, a003558 (2011).

43. Bley,C. J.et al.RNA–proteinbinding interface in the telomerase ribonucleoprotein.Proc. Natl Acad. Sci. USA 108, 20333–20338 (2011).

44. Lee, J., Mandell, E. K., Rao, T., Wuttke, D. S. & Lundblad, V. Investigating the role ofthe Est3 protein in yeast telomere replication. Nucleic Acids Res. 38, 2279–2290(2010).

45. Yen, W. F., Chico, L., Lei, M. & Lue, N. F. Telomerase regulatory subunit Est3 in twoCandida species physically interacts with the TEN domain of TERT and telomericDNA. Proc. Natl Acad. Sci. USA 108, 20370–20375 (2011).

46. Talley, J. M., DeZwaan, D. C., Maness, L. D., Freeman, B. C. & Friedman, K. L.Stimulation of yeast telomerase activity by the ever shorter telomere 3 (Est3)subunit is dependent on direct interaction with the catalytic protein Est2. J. Biol.Chem. 286, 26431–26439 (2011).

47. Zaug, A. J., Podell, E. R., Nandakumar, J. & Cech, T. R. Functional interactionbetween telomere protein TPP1 and telomerase. Genes Dev. 24, 613–622 (2010).

48. Sexton, A. N., Youmans, D. T. & Collins, K. Specificity requirements for humantelomere protein interaction with telomerase holoenzyme. J. Biol. Chem. 287,34455–34464 (2012).

Supplementary Information is available in the online version of the paper.

Acknowledgements This work was supported by grants from NSF MCB1022379 andNIH GM48123 to J.F., NIH GM54198 to K.C., GM071940 and AI069015 to Z.H.Z., RuthL. Kirschstein NRSA postdoctoral fellowship GM101874 to E.J.M., and RuthL. Kirschstein NRSA pre-doctoral training grant GM007185 fellowship for H.C. andD.D.C. We acknowledge the use of instruments at the Electron Imaging Center forNanoMachines supported by NIH (1S10RR23057 to ZHZ) and CNSI at UCLA.

Author Contributions J.J. and E.J.M. purified and characterized electron microscopysamples, collected and analysed electron microscopy data, and wrote the paper; K.H.,B.E. and B.M. designed and made strains, expression plasmids, initial purifications andreconstituted holoenzyme; H.C. purified telomerase; D.D.C. refined and modelledelements of TER; Z.H.Z., K.C. and J.F. supervised the research, analysed data and wrotethe paper.

Author Information Reprints and permissions information is available atwww.nature.com/reprints. The authors declare no competing financial interests.Readers are welcome to comment on the online version of the paper. Correspondenceand requests for materials should be addressed to J.F. ([email protected]), K.C.([email protected]) or Z.H.Z ([email protected]).

RESEARCH ARTICLE

6 | N A T U R E | V O L 0 0 0 | 0 0 M O N T H 2 0 1 3





www.nature.com/reprints


mailto:[email protected]



METHODSStrain construction, growth and analysis. Tetrahymena strains TERT–FZZ,Teb1–FZZ, p75–FZZ, p65–FZZ, p50–FZZ, MS2hp, p45–FZZ and p19–FZZ werepreviously described6,23. New strains ZZF–p50, TERT–ZZ/F–Teb1C and TERT–ZZ/F–Teb1BC were selected for targeted replacement of the endogenous locus(ZZF–p50 or TERT–ZZ) and/or for MTT1 promoter-driven transgene integ-ration at BTU1 (F–Teb1C, F–Teb1BC) using the neo2 and bsr2 cassettes6,49.Cells were grown to mid-log phase (2–4 3 105 cells per ml) at 30 uC in modifiedNeff medium (0.25% proteose peptone (EMD Milipore Chemicals), 0.25% BDBacto yeast extract (Becton Dickinson), 0.2% glucose, 30mM FeCl3). For large-scale purifications, transgene expression was induced by addition of 6 mM CdCl21 h before harvesting. Cell extract preparation and affinity purifications for sub-unit identification and activity assays were performed as described6. Genotypeswere verified by Southern blots and western blots using rabbit IgG (Sigma) todetect the ZZ tag or anti-Flag M2 mouse monoclonal antibody (Sigma) to detectthe Flag tag. Reconstituted enzyme activity assays were performed at room tem-perature using purified telomerase complexes on Flag antibody resin followingstandard Tetrahymena holoenzyme reaction conditions (50 mM Tris-acetate,pH 8.0, 10 mM spermidine, 5 mM 2-mercaptoethanol, 2 mM MgCl2, 100 nMd(GT2G3)3, 200mM dTTP, and 0.3mM 32P-labelled dGTP) for 5 min or with3.0mM 32P-labelled dGTP for 15 min6. Holoenzyme reconstitution used syntheticgenes encoding TERT–FZZ, p75, p65, p50, p45 and p19 for expression in RRL;TER purified following in vitro transcription by T7 RNA polymerase; andN-terminally His6-tagged Teb1BC purified following bacterial expression21.Cell extract preparation and affinity purification for electron microscopy.Cells were collected by centrifugation (5 min, 2,100g), washed in 20 mMHEPES NaOH, pH 8.0, and then resuspended in lysis buffer (30 ml per l ofgrowth) containing H2EG50 (20 mM HEPES NaOH, pH 8.0, 50 mM NaCl,1 mM EDTA TS, 1 mM TCEP HCl, and 10% glycerol) supplemented with 0.1%Triton X-100, 0.2% IGEPAL CA-630, 0.1 mM PMSF, 5mM proteosome inhibitorMG-132, and 1,000-fold dilution of Sigma Protease Inhibitor Cocktail P8340. Toeffect cell lysis, resuspensions were rotated end-over-end for 20 min and clearedby ultracentrifugation (1 h, 145,000g).

To minimize proteolysis, all purification steps were carried out at 4 uC andadditional washes were used in the early steps of purification as described below.Rabbit-IgG agarose slurry (Sigma) (3 ml per ml of cell extract) was washed 33 inlysis buffer, then added to cell extract for binding with end-over-end rotationovernight. Resin was collected by centrifugation (1 min, 3,200g) and washed 33

with no incubation and then 33 with end-over-end rotation (10 min) in washbuffer (20 mM HEPES NaOH, pH 8.0, 50 mM NaCl, 1 mM MgCl2, 1 mM TCEPHCl, 10% glycerol, and 0.1% IGEPAL CA-630). Telomerase was eluted withtobacco etch virus (TEV) protease (30 nM) in wash buffer for 1 h. After elution,supernatant containing holoenzyme was incubated with washed anti-Flag M2affinity gel (Sigma) for 1 h (4ml slurry per g of pelleted cells). After binding, anti-Flag M2 affinity gel was washed as above. Elution was effected by addition ofglycerol-free wash buffer containing 33Flag peptide (Sigma, 200 ngml21).Fab labelling for subunit identification. Anti-Flag Fab was prepared by incub-ating monoclonal mouse anti-Flag M2 IgG (Sigma) with resin-immobilizedpapain (Thermo Scientific Pierce) and purified according to the manufacturer’sprotocol. To prevent masking of the 33Flag epitope present on the target subunitby the 33Flag peptide required for elution, the tandem purification was done inreverse order with anti-Flag M2 affinity gel purification followed by rabbit-IgGagarose purification. Washing and elution were done as described above. To effectFab labelling, holoenzyme-enriched rabbit-IgG agarose was incubated (1 h, 4 uC)with wash buffer containing anti-Flag Fab (7.5 mg ml21). Excess Fab was removedby washing 53 with wash buffer. Labelled holoenzyme was eluted with washbuffer containing His6-TEV protease (2 nM). After elution, His6-TEV proteasewas removed from the elution by incubating with washed Ni-NTA agarose(30 min, 4 uC).Holoenzyme purification by MS2cp. Tetrahymena expressing MS2hp TER werecollected, lysed and clarified as described above. N-terminal His6–ZZF–MS2cpbacterially expressed from pET28 was purified using Ni-NTA agarose. Rabbit-IgG agarose was incubated with His6–ZZF–MS2cp in a 1:1 stoichiometry toanticipated holoenzyme yield based on previous preparations. The enriched rab-bit-IgG agarose was then washed 33 and added to clarified extract for overnightincubation. Purification then proceeded as described above.Electron microscopy specimen preparation and data collection. For negativestaining electron microscopy, 2ml of telomerase sample was applied to a glow-discharged grid coated with carbon film. The sample was left on the carbon film for10 s, followed by negative staining with 0.8% uranyl formate. Negatively stainedspecimens were carefully examined by electron microscopy, and only the regionswith particles fully embedded in stain were selected for imaging. Electron micro-scopy micrographs were recorded on a TIETZ F415MP 16-megapixel CCD

camera at 68,0273 magnification in an FEI Tecnai F20 electron microscope oper-ated at 200 kV. The micrographs were saved by 23 binning and had a pixel size of4.4 A. To overcome the problems of preferred orientation and structural variabilityof the samples, 3D reconstructions were carried out using RCT method, whichtakes advantage of classification to differentiate different conformations and tiltingto get lateral views for 3D reconstruction22. RCT images were collected with thegrids tilted at two angles successively (65u and 0u) for each specimen area of interestwith the assistance of Leginon automation software50,51. To minimize the potentialsmearing/stretching on 3D maps due to the missing cone problem in RCT datacollection, a high tilt angle (65u) was used. For samples where only 2D imageanalysis was performed, micrographs were taken without tilting of the grids.Total numbers of micrographs and particles used for analysis are summarizedin the Supplementary Table.

For cryo-electron microscopy, 2.5ml of sample was applied to a glow-dis-charged Quantifoil R2/1 grid. The grid was then blotted with filter paper toremove excess sample, and flash-frozen in liquid ethane with an FEI VitrobotMark IV. The grid was loaded into an FEI Titan Krios electron microscopeoperated at 120 kV. Micrographs were acquired with a Gatan UltraScan400016-megapixel CCD camera at 65,1623 magnification with defocus values rangingfrom 21.9 to 24.0 mm and an exposure dose of 20 e2 A22, and without tilting ofthe grids. The micrographs with a pixel size of 2.3 A were used for data processingwithout binning.2D image reference-free classification. Particles from the untilted micrographswere automatically picked using DoGpicker52. All particles in the micrographs,including a small number of aggregated particles, were initially picked to avoidbias in selecting particles. Particles were boxed out in 96 3 96 pixels (or 144 3 144pixels for cryo-electron microscopy images) using batchboxer in EMAN53. Thedefocus value of each micrograph was determined by CTFFIND54 and the part-icles were corrected for contrast transfer function (CTF) by phase-flipping withthe corresponding defocus and astigmatism values using bsoft55. The phase-cor-rected particles were then iteratively (normally 9 iterations) classified using therefine2d.py program in EMAN. The classification for each sample was doneindependently without using any model or reference. The aggregated particlesclustered in the few classes showing poor structures were manually eliminatedduring the iterations of classification. For the samples at very low concentration(Fab-labelled p75–F, p19–F and p65–F telomerase, and MS2cp-labelled MS2hptelomerase), a looser criterion for DoGpicker was used to pick more particles andconsequently some ‘fake’ particles from the stain background were also selected.The class averages were visually inspected at several different iterations of clas-sification to remove the ‘fake’ particles that were clustered in a small set of classaverages that show no structural features.

For the RCT 3D reconstructions (Supplementary Table), the screened particlesfrom the above steps were re-classified using the Correspondence Analysismethod in SPIDER56.The class averages from the above EMAN classification stepwere used as references for multi-reference alignment in SPIDER. Each new classfrom this analysis was used for an RCT 3D reconstruction.RCT 3D reconstruction and refinement. Corresponding particles from tilt pairs(65u and 0u) were picked using ApTiltPicker.py in Appion52 modified to optimizeautomatic particle picking and verified by visual inspection. The 0u tilt particleswere classified as described above. Next, 3D RCT maps were reconstructed from65u tilt particles. Particles within each class of the 0u tilt data set all have the sameconformation and the same orientation except in-plane rotation. Their corres-ponding 65u tilt particles have different orientations due to in-plane rotation andwere combined to reconstruct one 3D map. Each 3D map was iteratively refinedwith SPIDER followed by FREALIGN57 by refinement of the centre of 65u tiltparticles in the corresponding class. A representative refinement procedure isshown in Supplementary Fig. 3b. The orientations of 65u tilt particles were cal-culated from the in-plane rotation of the corresponding 0u tilt particles and thegeometric relationship between the 65u and 0u tilt micrographs, and kept fixedduring structure refinement. Multiple 3D maps were generated from RCT 3Dreconstruction because particles in different conformations were classified intodifferent classes each of which generated an individual 3D map. Representative3D maps are shown in Supplementary Fig. 3c. Each 3D map was refined usingonly the 65u tilt particles in its corresponding class. The defocus value of each 65utilt particle was calculated by CTFTILT program54 and then applied to 3D recon-structions with FREALIGN for proper correction of phase and amplitude in CTF.

To estimate the resolution of the representative 3D map of Teb1–F telomerasein the ‘stable’ conformation, the 2,220 65u tilt particles used to reconstruct that 3Dmap were split into odd- and even-numbered halves and the halves of data wereused to reconstruct two 3D maps with FREALIGN using the determined orienta-tions and image centres, respectively, to calculate Fourier shell correlation (FSC).The resolution is 25 A at FSC 5 0.5 (Supplementary Fig. 4b). For comparison, theFSC was also calculated using the recently introduced ‘gold standard’ FSC

ARTICLE RESEARCH


method58, in which the halves of data (both 65u and 0u tilt) were split at thebeginning for independent orientation determination, 3D reconstruction, andrefinement to generate two independent 3D maps. The ‘gold standard’ FSCsuggests resolutions of 29 A at FSC 5 0.5 and 24 A at FSC 5 0.143(Supplementary Fig. 4b). Since a high tilt angle (65u) was used for RCT 3Dreconstructions, the missing cone was minimized and did not have notableimpact on the structural interpretation of the 3D maps at the present resolution.Evaluation of sample integrity by size-exclusion chromatography (SEC). Toconfirm that the affinity-purified telomerase holoenzymes were intact after puri-fication and that negative staining did not induce significant aggregation, weobtained an SEC profile of a sample of affinity purified TERT–F telomeraseand electron microscopy images from the peak fractions (Supplementary Fig.1c, d). The sample (50ml) was run at 50ml min21 on a Superdex 200 column inthe same buffer used for the electron microscopy samples except IGEPAL CA-630was replaced by Tween-80. The SEC profile shows that the telomerase holoen-zyme runs as a single peak with a constant A260 nm/A280 nm ratio indicative of alargely intact RNA–protein complex. The class averages of particles from the peakfraction are identical to those for the affinity-purified sample (data not shown).There are no later eluting peaks from dissociated subunits (which would havedifferent A260 nm/A280 nm ratios).Localization of Fab-labelled subunits. Each protein was assigned by identifyingthe characteristic Fab density in the class averages and 3D reconstructions. The33Flag tag allows up to three Fabs to bind simultaneously. Fab was observedattached to all tagged telomerase particles except for p45–F. For p45–F telomer-ase, Fabs were observed bound to free p45–F protein, while the p45–F telomeraseholoenzyme particles appeared unchanged. The location of p45 in the holoen-zyme was therefore assigned based on the density remaining after assignment ofTERT–F, Teb1–F, p75–F, p65–F, F–p50 and p19–F.Fitting of atomic models into the 3D electron microscopy maps. The homologymodel of Tetrahymena TERT RT-CTE domains was built using amino acids 520–1117 and the Tribolium TERT crystal structure (PDB ID: 3KYL) as a templatewith SWISS-MODEL27. The model of Tetrahymena TERT (TRBD, RT, and CTEdomains) was built by joining the crystal structure of Tetrahymena TRBD (PDBID: 2R4G) and the homology model of RT-CTE domains using the TriboliumTERT crystal structure as reference for structural alignment. The model ofTetrahymena TERT was then manually placed in an approximate position inthe electron microscopy density map, followed by docking into the electronmicroscopy density map accurately using the ‘‘Fit in Map’’ function in UCSFChimera59. The location of the TEN domain, absent in Tribolium TERT17, wassupported by definition of the subunit boundaries of the adjacent Teb1C and p50.The crystal structure of Tetrahymena TERT TEN domain (PDB ID: 2B2A) wasmanually placed into the electron microscopy density map. Its position wasiteratively adjusted to fit the empty density adjacent to the boundaries of Teb1and p50 and the rest of TERT. Because the orientation of the TEN domain couldnot be determined based on the electron microscopy density due to the limit ofresolution, we oriented it to be consistent with the homology model of humanTERT28. For fitting of p65 xRRM2:SL4, the putative electron microscopy densityof p65–TER cut from the full electron microscopy density map was used in orderto avoid interference from unrelated structures. The model of p65 xRRM2:SL48

was manually placed in the electron microscopy density map based on the loca-tion of the C terminus determined by Fab labelling and then docked using the ‘‘Fitin Map’’ function in UCSF Chimera. Its position was further manually adjusted toavoid clashing with the model of TERT. To confirm the fittings of TERT and p65xRRM2:SL4, these two structures were also fit into the electron microscopydensity map using the colores program in Situs60. The fitting with colores wasautomated by 6D-space search and contour-based matching, and did not requiremanual placing of high-resolution structures to approximate positions in electronmicroscopy density maps. The crystal structure of Teb1C (PDB ID: 3U50) wasmanually placed into the electron microscopy density map and its orientation wasadjusted so that the position of its C terminus was consistent with the result fromthe Fab labelling experiment. The homology model of p65 La-RRM1:AUUUU-398,61 was manually placed into the electron microscopy density map, and itsorientation was adjusted to match the shape of the electron microscopy densitymap. Cross-correlation for the fittings were: TERT TRBD-RT-CTE, 0.91; TERTTEN, 0.84; p65 xRRM2:SL4, 0.80; p65 La-RRM1:AUUUU-39, 0.86; and Teb1C,0.82. Although the position and orientation of p65 xRRM2:SL4 are well deter-mined, the cross-correlation coefficient reflects the absence of the b2–b3 loop andamino acids linking xRRM2 to RRM1 in the crystal structure8.Structures of SL2, distal SL4 and the PK. Previously reported NMR structures ofSL2 and distal SL425,26 were re-refined using a new nuclear Overhauser enhancement(NOE) restraint list derived from re-analysis of all previously obtained 2D nuclearOverhauser enhancement spectroscopy (NOESY) and 3D NOESY-heteronuclearmultiple quantum correlation (HMQC) spectra and a new set of residual dipolar

couplings (RDC) restraints (63 for SL2 and 48 for distal SL4) collected using pf1bacteriophage for the alignment media. The new PDB accession numbers for SL2and distal SL4 are 2M22 and 2M21, respectively, and BMRB accession numbers are18892 and 18891, respectively. A grid search produced optimal values for the mag-nitude and asymmetry of the RDC alignment tensor of Da 5 222.3 Hz, R 5 0.38 forSL2, and Da 5 217.5 Hz, R 5 0.40 for SL4. The structures were calculated from anextended, unfolded RNA conformation using XPLOR-NIH v.2.9.862,63 followingstandard XPLOR protocols. A model structure of the PK was generated based onthe secondary structure and two U-A-U triples predicted by sequence analysis64.XPLOR-NIH was used to calculate the structure from an extended RNA conforma-tion as previously described62,63. A mock NMR restraint list was derived based on thesolution structure of the human telomerase PK65. For the two stems, A-form RNArestraints were used for the dihedral angles and NOEs. NOEs for residues in thestems include H6/H8 (Py/Pu) to its own sugar protons and to the sugar protons(H19, H29, H39, H49) of the n 2 1 residue, H6/H8 to H6/H8 sequential, and iminosequential NOEs. For the PK loop 1, base-triple hydrogen bond restraints, planar-ity restraints with PK stem 2 residues, and sequential stacking NOEs were used toform the structured triple helix. PK loop 2 was given H6/H8 sequential NOEs andH6/H8-H19 NOEs within the minor groove of PK stem 1 to place it in the expectedconformation.Modelling of TER. First, the position of the template in TER was determined bymatching the crystal structure of Tribolium TERT-nucleic acid complex (PDB ID:3KYL) with the docked model of Tetrahymena TERT. Second, the position of SL4was determined by the docking of the atomic model of p65 xRRM2:SL4 complexinto the electron microscopy density map. Third, the atomic model of SL2 wasmanually placed into the electron microscopy density map and its position wasiteratively adjusted to match the result from MS2cp labelling experiment. S1(modelled as A-form RNA) and the PK were placed into the electron microscopydensity map based on physical constraints with the rest of TER. S1 is connected toS4 by 4 nucleotides and to S2 by 10 nucleotides. The model of the 31 nucleotidePK, which is connected to S1 by 3 nucleotides and to the template by the TRE, wasfit into the only region of unoccupied density, between the TRBD and TENdomains, remaining after all the other major elements of TERT and TER werelocalized. Although the modelled positions of the TRE 39 to the template (Fig. 1a)and the top of the pseudoknot could potentially be swapped with the TENdomain, they would still be on the same side (right side or behind in Fig. 2a or2b, respectively) of the TERT TRBD-RT-CTE ring that has the active site andbound template. In either case, the TRE appears to be close to the TEN domain.The single-stranded regions of TER (except the template) were initially modelledas ideal A-form single-stranded RNA and their structures were modified usingthe ‘‘Minimize structure’’ function in UCSF Chimera to connect with theirrespective folded RNA fragments. The electron microscopy density maps andthe modelled protein and RNA structures were visualized with UCSF Chimera.

49. Couvillion,M.T.&Collins,K.Biochemical approaches including thedesignanduseofstrains expressing epitope-tagged proteins. Methods Cell Biol. 109, 347–355 (2012).

50. Suloway, C. et al. Automated molecular microscopy: The new Leginon system.J. Struct. Biol. 151, 41–60 (2005).

51. Suloway, C. et al. Fully automated, sequential tilt-series acquisition with Leginon.J. Struct. Biol. 167, 11–18 (2009).

52. Voss, N. R., Yoshioka, C. K., Radermacher, M., Potter, C. S. & Carragher, B. DoGPicker and TiltPicker: Software tools to facilitate particle selection in single particleelectron microscopy. J. Struct. Biol. 166, 205–213 (2009).

53. Ludtke, S. J., Baldwin, P. R. & Chiu, W. EMAN: Semiautomated software for high-resolution single-particle reconstructions. J. Struct. Biol. 128, 82–97 (1999).

54. Mindell, J. A.&Grigorieff,N.Accuratedeterminationof localdefocusandspecimentilt in electron microscopy. J. Struct. Biol. 142, 334–347 (2003).

55. Heymann, J. B. Bsoft: Image and molecular processing in electron microscopy.J. Struct. Biol. 133, 156–169 (2001).

56. Frank, J. et al. SPIDER and WEB: Processing and visualization of images in 3Delectron microscopy and related fields. J. Struct. Biol. 116, 190–199 (1996).

57. Grigorieff, N. FREALIGN: High-resolution refinement of single particle structures.J. Struct. Biol. 157, 117–125 (2007).

58. Scheres, S. H. & Chen, S. Prevention of overfitting in cryo-EM structuredetermination. Nature Methods 9, 853–854 (2012).

59. Pettersen, E. F. et al. UCSF Chimera—A visualization system for exploratoryresearch and analysis. J. Comput. Chem. 25, 1605–1612 (2004).

60. Chacon, P. & Wriggers, W. Multi-resolution contour-based fitting ofmacromolecular structures. J. Mol. Biol. 317, 375–384 (2002).

61. Jacks, A. et al. Structure of the C-terminal domain of human La protein reveals anovel RNA recognition motif coupled to a helical nuclear retention element.Structure 11, 833–843 (2003).

62. Schwieters,C.D., Kuszewski, J. J.&MariusClore,G.UsingXplor–NIHforNMRmolecularstructure determination. Prog. Nucl. Magn. Reson. Spectrosc. 48, 47–62 (2006).

63. Schwieters, C. D., Kuszewski, J. J., Tjandra, N. & Marius Clore, G. The Xplor–NIHNMR molecular structure determination package. J. Magn. Reson. 160, 65–73(2003).

RESEARCH ARTICLE


64. Ulyanov, N. B., Shefer, K., James, T. L. & Tzfati, Y. Pseudoknot structures withconserved base triples in telomerase RNAs of ciliates. Nucleic Acids Res. 35,6150–6160 (2007).

65. Theimer, C. A., Blois, C. A. & Feigon, J. Structure of the human telomerase RNApseudoknot reveals conserved tertiary interactions essential for function. Mol. Cell17, 671–682 (2005).

ARTICLE RESEARCH


W W W. N A T U R E . C O M / N A T U R E | 1

SUPPLEMENTARY INFORMATIONdoi:10.1038/nature12062

a bSupplementary Figure 1

c

Supplementary Figure 1. EM micrographs of telomerase and characterization of sample integrity. a, Negative-staining EM micrograph of TERT-f telomerase with 12 representative particles picked (red boxes) and shown on the top. The corresponding class averages are shown side by side for comparison. b, CryoEM micrograph of TERT-f telomerase. Twelve picked particles are shown on the top. Negative staining class averages in similar view as the cryoEM particle images are shown side by side for comparison, with the contrast inverted to help highlight the structural features in the cryoEM particle images. c, Size-exclusion Superdex 200 chromatography profile (UV traces: blue, 260 nm; red, 280 nm) of affinity purified TERT-f telomerase. Gel filtration standards (BIO-RAD 151-1901) are shown in green with molecular masses indicated. The telomerase holoenzyme runs as a single peak with a constant A260/A280 ratio indicative of an intact RNA-protein complex and there are no later eluting peaks from dissociated subunits (which would have different A260/A280 ratios). The small shoulder at the beginning of the telomerase peak is due to a small amount of aggregated particles, as confirmed by EM (data not shown). d, Negative stain-ing EM image of peak fraction A4, which shows particles with identical features as those obtained for samples without SEC (a), and without the small number of aggregates which were eliminated during 2D image classification for the EM samples analyzed in this study.

Particles

Classaverages

Particles

Classaverages

Particles

Classaverages

Particles

Classaverages

50nm 50nm

0.0

(mAu)

A13A1 A2 A3 A4 A5 A6 A7 A8 A9 A10 A11 A12

10.0 20.0 30.0 40.0 (min)

670 kDa 158 kDa 44 kDa 17 kDa 1.35 kDa

TERT-fTelomeraseHoloenzyme

3×FLAGPeptide

0

60

40

20

50nm

d

SUPPLEMENTARY INFORMATION

2 | W W W. N A T U R E . C O M / N A T U R E

RESEARCH

a b c d

31% 5% 21% 34%

e

f

Supplementary Figure 2TERT-f TERT-f (cryoEM) Teb1-f f-p50 p75-f

Supplementary Figure 2. 2D image classification of telomerase. a-e, Representative class averages of negative-staining EM (a, c-e) and cryoEM (b) images of TERT-f (a, b), Teb1-f (c), f-p50 (d), and p75-f (e) telomerase. Comparison of the class averages of TERT-f telomerase from negative-staining EM (a) and cryoEM (b) shows that they have the same structural features, as well as preferred orientation and multiple conformations. Note that the density of Teb1 is present in all particles of Teb1-f telomerase (c) and is absent in all particles of f-p50 telomerase (d), while TERT-f telomerase (a, b) has both of these particles (with and without Teb1). The class averages of p75-f telomerase (e), which must have stoichio-metric p75, show the multiple positions of the 7-1-4 subcomplex seen in the other tagged telomerase samples (a-d). f, Dendrogram of hierarchically clustered class averages of TERT-f telomerase generated by SPIDER. The estimated percentage of particles in the different conformations is: 31% in “stable” conformation; 5% lacking Teb1; 21% lacking p65 N-terminal region; and 34% with 7-1-4 in different posi-tions. The numbers of particle images used to calculate the class averages are displayed inside (a-e) or above (f) each class average. g, Comparison of class averages with raw particles from TERT-f telomer-ase. Each column shows a class average and five representative raw particles aligned to the correspond-ing class average.

150 170 159

110 299 256

172 136 166

169 130 161

245 318 338

279 379 247

244 98 138

43 44 203

771 561

981 791

539 615

458 473

5457 5798 4263

3574 4278 5056

2164 2261 1465

3721 3267 1930

231 168 256

134 258 374

105 84 249

430 349 359

Classaverages

Rawparticles

g


SUPPLEMENTARY INFORMATION RESEARCH

Supplementary Figure 3

Supplementary Figure 3. RCT data collection, 3D reconstructions, and structure refinement. a, Tilt pair (0° and 65°) of micrographs of a negatively stained specimen of TERT-f telomerase. Raw images are overlaid with red crosses marking the corresponding particles picked automatically from the pair of micro-graphs. To avoid bias in selecting particles, all particles were initially picked. The few aggregated particles were eliminated during image classification. b, Iterative refinement of RCT 3D reconstructions of Teb1-f telomerase. 3D maps were iteratively reconstructed and refined with 2,220 tilted particle images using SPIDER (iter 1-7, in which only the phase of CTF was corrected) and FREALIGN (iter 8, in which the phase and amplitude of CTF were both corrected. We note that the black border around the particle in iter 8, which is from CTF artifacts, disappeared after correction of amplitude.) The projections (φ=0, θ=0, ψ=0, which is also the orientation of the untilted class average) of the 3D maps are in good agreement with the class average of the corresponding untilted particle images that was generated by Correspondence Analysis in SPIDER. We note that the class average is from untilted images and the 3D maps are from tilted images, respectively, and the consistency between them indicates the reliability of our RCT 3D reconstructions. Four views of each 3D map from the RCT reconstructions are presented (iterations 1-8), demonstrating the progressive improvement of structure. c, Representative 3D maps from RCT 3D recon-structions of Teb1-f telomerase. Since these particles were isolated from the Teb1-fzz strain, all of the particles must have Teb1. Consistent with this, all of the 3D maps show Teb1C. Differences in the maps are due to the positional flexibility of the 7-1-4 subcomplex and to the N-terminal truncation of p65 for some maps (vi, vii, and x). In some 3D maps the position of 7-1-4 is too flexible to be visible (viii, ix, and x), but 7-1-4 is present as fuzzy density in the class averages.

Cla

ss a

vera

ge(u

ntilt

ed)

iter 1

iter 2

iter 3

iter 4

iter 5

iter 6

iter 7

iter 8(FREALIGN)

Pro

ject

ions

of R

CT

reco

nstru

ctio

ns

90°

65°

0°

100nm

i

ii

iii

iv

v

vi

vii

viii

ix

x

a b c90° 90°

90° 90° 90°

3D m

aps

of R

CT

reco

nstru

ctio

ns



RESEARCH

a bTERT-f Teb1-f f-Teb1C

90°

90°

90°

Spatial Frequency (1/Å)

Four

ier S

hell

Cor

rela

tion


Supplementary Figure 4. Comparison of 3D reconstructions of telomerase. a, Four views of the 3D reconstructions of (left to right) TERT-f, Teb1-f, and f-Teb1C telomerases. Comparison shows they are highly similar, suggesting that only the C-terminal domain of Teb1 (red circles) is visible in the EM recon-structions, and the other domains of Teb1 when present are positionally flexible. b, Fourier shell correla-tion (FSC) of the RCT 3D reconstruction of 2,220 particles from Teb1-f telomerase holoenzyme in the “stable” conformation between odd- and even-numbered particles. For comparison, the FSC was calcu-lated using a conventional method (black line), in which the halves of data were spilt before the last round of 3D reconstruction, and using the recently introduced “gold standard” FSC method (red line), in which the halves of data were split at the beginning for independent data processing. The conventional FSC suggests a resolution of 25 Å at 0.5 FSC criterion and the “gold standard” FSC suggests resolutions of 29 Å at FSC=0.5 and 24 Å at FSC=0.143.

100 75 50 250.0

0.5

1.0

(29 Å, FSC=0.5)

(24 Å, FSC=0.143)

(25 Å, FSC=0.5)



a

c

b

d e f

90°

90°

C

C C C

CCCCCA

A A A A AA

AAAA U

U U U

UUUU

G G G G G G

GGG205′

3′

MS2hp stem 2g


90°

90°

90°

90°

TERT-f-Fab

Fab-f-p50

Teb1-f-Fab

p65-f-Fab p75-f-Fab p19-f-Fab

MS2hp-MS2cp

Supplementary Figure 5. Class averages and 3D reconstruction of Fab-labeled and MS2cp-labeled telomerase. a-c, Class averages (left) and 3D reconstructions (right) of TERT-f-Fab (a), Teb1-f-Fab (b), and Fab-f-p50 (c) telomerase. The density of Fab is highlighted with yellow color in the 3D reconstructions. d-f, Class averages of p65-f-Fab (d), p75-f-Fab (e), and p19-f-Fab (f) telomerase. Note that Fab is bound to particles lacking the N-terminal part of p65-f (second row in d) as well as to the full-length p65-f. Up to 3 Fabs are seen in the particles of Fab-labeled telomerase because Fabs can bind to each of the three FLAG epitopes. g, Class averages of MS2cp-labeled MS2hp telomerase and schematic of MS2hp substitution of the loop of stem-loop 2 (red nucleotides). The binding of MS2cp (white arrow) appears to displace p75 (black arrow) from the stable position.



RESEARCH Supplementary Figure 6

IgG then FLAG IP IgG then FLAG IP,MNase before 2nd elution

85100

70

120

60

50

40

30

kDaMoc

k

p45-

fzz

p50-

fzz

zzf-p

50

Moc

k

p45-

fzz

p50-

fzz

zzf-p

50

150200

2520

15

TERT

p75, p65

f-p50

p45p45-f, p50

p19

o o

o ooo

ooo

o

o o o o

o o

oo

o

**

* *

**

f-p50p50

p45-f

oo

oo

5543

5543

FLAG Ab:

p50 C-terminus Ab:

1 2 3 4 5 6 7 8

Supplementary Figure 6. Comparison of tagged p50 and p45 purifications. Tandem affinity purifica-tion was performed from cell extracts containing no tagged protein, either p45-fzz or p50-fzz expressed in replacement of the endogenous locus, or zzf-p50 expressed in partial replacement of the endogenous locus. Prior to the second-step 3×FLAG peptide elution, antibody resin was divided and incubated with or without micrococcal nuclease (MNase), followed by extensive washes. Aliquots of the same sample were used for SDS-PAGE and silver staining (upper panel) or for western blots with affinity purified rabbit polyclonal antibody against the p50 C-terminal peptide or triple FLAG tag (lower panels). The migration positions of full-length holoenzyme proteins are indicated by an open circle, while asterisks indicate likely products of partial proteolysis. Using equivalent amounts of input cell extract, p45-fzz purifies a higher yield of holoenzyme subunits than zzf-p50, but zzf-p50 purifies a much higher yield of holoenzyme subunits than p50-fzz. MNase digestion removes TER, p65, and TERT from p45-f or f-p50 and removes some p50 from p45-f (compare the p45-f and p50 western blot signals in lanes 2 and 6), but some p50 remains associated with 7-1-4 in the absence of the RNP catalytic core. Note that untagged p50 co-migrates with p45-f.



a

b

c

d


-180 -120 -60 0 60 120 1800.7

0.8

0.9

1.0

Cro

ss-c

orre

latio

n

-180 -120 -60 0 60 120 1800.6

0.7

0.8

0.9C

ross

-cor

rela

tion

φ

θ

θ

Supplementary Figure 7. Fitting of TERT model (TRBD-RT-CTE) and p65 xRRM2:stem-loop 4 into the EM density map using UCSF Chimera and Situs. a, Comparison of Tetrahymena TERT models fit into the EM density map (grey) using UCSF Chimera (blue) and Situs (yellow), showing their consistency with rmsd=3.9 Å. b, Cross correlation between the EM density map and the TERT model. The TERTmodel is rotated from the best fit orientation (φ=0°, θ=0°) along two axes indicated in a. Since the donut- like shape of TERT is an unambiguous constraint for fit orientation, the change of cross correlation is demonstrated by in-plane rotation (θ) and flipping (φ). c, Comparison of p65 xRRM2:stem-loop 4 fit into the putative EM density of p65:TER (grey mesh), which is cut from the full density map (grey surface), using UCSF Chimera (orange) and Situs (yellow), and with manual adjustment of position (green) to avoid clash with TERT (blue). UCSF Chimera and Situs both fit p65 xRRM2:stem-loop 4 to the same orientation with a small difference of translation (rmsd=8.6 Å). The manual adjustment includes a translation and a small tilting with rmsd=14.0 Å between green and yellow models. d, Cross correlation between the EM density map and p65 xRRM2:stem-loop 4. The rotation of p65 xRRM2:stem-loop 4 (green in c, θ=0°) is along the axis indicated in c, while rotations along other axes result in a greater drop of cross correlation.n



RESEARCH


Teb17-1-4–

–++

+– –

–

––

–

–

+

++

+

+ ++

++–

––

RC

p50

1 2 3 4 5 6 7 8

Supplementary Figure 8. High RAP reconstitution by recombinant subunits. The RNP catalytic core alone (TERT-TER-p65) was reconstituted with additional combinations of 7-1-4, p50, and/or Teb1BC and assayed for repeat synthesis in 15 min reactions with 3 μM dGTP. RC is a radiolabeled oligonucleotide added to telomerase products as a precipitation recovery control.



SUPPLEMENTARY TABLE | Summary of strains and EM micrographs used in the EM

studies

Unlabeled Fab or MS2cp labeled

Untilted RCT Untilted RCT

Tagged subunits

Micro-graph count

Particle count

Micro-graph pair count

Particle count

Micro-graph count

Particle count

Micro-graph pair count

Particle count

TERT-fzz 4,043 422,347 316 11,816 5,651 152,000 522 11,004

TERT-fzz (cryoEM) 1,032 174,863

TERT-zz/f-Teb1C 4,778 363,981 430 4,371

TERT-zz/f-Teb1BC 4,649 153,508

Teb1-fzz 3,729 556,923 1,253 55,319 1,962 147,713 339 7,525

p75-fzz 2,161 476,602 3,084 17,918

p65-fzz 5,098 294,675 6,251 27,945

zzf-p50 2,610 52,984 6,368 60,779 802 8,029

MS2hp 6,207 5,646

p45-fzz 774 67,482 6,699 267,899

p19-fzz 4,812 1,589

the architecture of tetrahymena telomerase...

Documents